# We don't have access to install libraries in the workspace
# Please make sure those are installed
!pip install -q -r ./requirements.txt

import os
os.environ['OPENAI_API_KEY'] = "<YOUR API KEY>"

from langchain.llms import OpenAI

model_name = 'gpt-3.5-turbo'
llm = OpenAI(model_name=model_name, temperature=0)

INSTRUCTION = "Generate fifteen realistic real estate listings from diverse neighborhoods."
SAMPLE_LISTING = \
"""
Here's a sample listing:

Neighborhood: Green Oaks
Price ($): 800,000
Bedrooms: 3
Bathrooms: 2
House Size (sqft): 2,000
Description: Welcome to this eco-friendly oasis nestled in the heart of Green Oaks. This charming 3-bedroom, 2-bathroom home boasts energy-efficient features such as solar panels and a well-insulated structure. Natural light floods the living spaces, highlighting the beautiful hardwood floors and eco-conscious finishes. The open-concept kitchen and dining area lead to a spacious backyard with a vegetable garden, perfect for the eco-conscious family. Embrace sustainable living without compromising on style in this Green Oaks gem.
Neighborhood Description: Green Oaks is a close-knit, environmentally-conscious community with access to organic grocery stores, community gardens, and bike paths. Take a stroll through the nearby Green Oaks Park or grab a cup of coffee at the cozy Green Bean Cafe. With easy access to public transportation and bike lanes, commuting is a breeze.
"""

C:\Users\roysa\.conda\envs\udacity-genai\Lib\site-packages\langchain_community\llms\openai.py:249: UserWarning: You are trying to use a chat model. This way of initializing it is no longer supported. Instead, please use: `from langchain_community.chat_models import ChatOpenAI`
  warnings.warn(
C:\Users\roysa\.conda\envs\udacity-genai\Lib\site-packages\langchain_community\llms\openai.py:1070: UserWarning: You are trying to use a chat model. This way of initializing it is no longer supported. Instead, please use: `from langchain_community.chat_models import ChatOpenAI`
  warnings.warn(

from langchain.output_parsers import PydanticOutputParser
from langchain_core.pydantic_v1 import BaseModel, Field, NonNegativeInt
from typing import List

class RealEstateListing(BaseModel):
    neighborhood: str = Field(description="Name of the neighborhood")
    price: NonNegativeInt = Field(description="Price of the property in USD")
    bedrooms: NonNegativeInt = Field(description="Number of bedrooms in the property")
    bathrooms: NonNegativeInt = Field(description="Number of bathrooms in the property")
    house_size: NonNegativeInt = Field(description="Size of the property in square feet")
    description: str = Field(description="Description of the property.")   
    neighborhood_description: str = Field(description="Description of the neighborhood.")  

class ListingCollection(BaseModel):
    listing: List[RealEstateListing] = Field(description="List of available real estate")
        
parser = PydanticOutputParser(pydantic_object=ListingCollection)

from langchain.prompts import PromptTemplate

prompt = PromptTemplate(
    template="{instruction}\n{sample}\n{format_instructions}\n",
    input_variables=["instruction", "sample"],
    partial_variables={"format_instructions": parser.get_format_instructions},
)
query = prompt.format(instruction = INSTRUCTION, sample = SAMPLE_LISTING)
print(query)

Generate fifteen realistic real estate listings from diverse neighborhoods.

Here's a sample listing:

Neighborhood: Green Oaks
Price ($): 800,000
Bedrooms: 3
Bathrooms: 2
House Size (sqft): 2,000
Description: Welcome to this eco-friendly oasis nestled in the heart of Green Oaks. This charming 3-bedroom, 2-bathroom home boasts energy-efficient features such as solar panels and a well-insulated structure. Natural light floods the living spaces, highlighting the beautiful hardwood floors and eco-conscious finishes. The open-concept kitchen and dining area lead to a spacious backyard with a vegetable garden, perfect for the eco-conscious family. Embrace sustainable living without compromising on style in this Green Oaks gem.
Neighborhood Description: Green Oaks is a close-knit, environmentally-conscious community with access to organic grocery stores, community gardens, and bike paths. Take a stroll through the nearby Green Oaks Park or grab a cup of coffee at the cozy Green Bean Cafe. With easy access to public transportation and bike lanes, commuting is a breeze.

The output should be formatted as a JSON instance that conforms to the JSON schema below.

As an example, for the schema {"properties": {"foo": {"title": "Foo", "description": "a list of strings", "type": "array", "items": {"type": "string"}}}, "required": ["foo"]}
the object {"foo": ["bar", "baz"]} is a well-formatted instance of the schema. The object {"properties": {"foo": ["bar", "baz"]}} is not well-formatted.

Here is the output schema:
```
{"properties": {"listing": {"title": "Listing", "description": "List of available real estate", "type": "array", "items": {"$ref": "#/definitions/RealEstateListing"}}}, "required": ["listing"], "definitions": {"RealEstateListing": {"title": "RealEstateListing", "type": "object", "properties": {"neighborhood": {"title": "Neighborhood", "description": "Name of the neighborhood", "type": "string"}, "price": {"title": "Price", "description": "Price of the property in USD", "minimum": 0, "type": "integer"}, "bedrooms": {"title": "Bedrooms", "description": "Number of bedrooms in the property", "minimum": 0, "type": "integer"}, "bathrooms": {"title": "Bathrooms", "description": "Number of bathrooms in the property", "minimum": 0, "type": "integer"}, "house_size": {"title": "House Size", "description": "Size of the property in square feet", "minimum": 0, "type": "integer"}, "description": {"title": "Description", "description": "Description of the property.", "type": "string"}, "neighborhood_description": {"title": "Neighborhood Description", "description": "Description of the neighborhood.", "type": "string"}}, "required": ["neighborhood", "price", "bedrooms", "bathrooms", "house_size", "description", "neighborhood_description"]}}}
```

output = llm(query)

C:\Users\roysa\.conda\envs\udacity-genai\Lib\site-packages\langchain_core\_api\deprecation.py:117: LangChainDeprecationWarning: The function `__call__` was deprecated in LangChain 0.1.7 and will be removed in 0.2.0. Use invoke instead.
  warn_deprecated(

from fastapi.encoders import jsonable_encoder
import pandas as pd

result = parser.parse(output)
df = pd.DataFrame(jsonable_encoder(result.listing))
df.head()

df.to_csv('listings.csv',index_label = 'id')

import pandas as pd

df = pd.read_csv('listings.csv')

from diffusers import DiffusionPipeline, AutoPipelineForText2Image
from diffusers.utils import load_image, make_image_grid
import torch

C:\Users\roysa\.conda\envs\udacity-genai\Lib\site-packages\diffusers\utils\outputs.py:63: UserWarning: torch.utils._pytree._register_pytree_node is deprecated. Please use torch.utils._pytree.register_pytree_node instead.
  torch.utils._pytree._register_pytree_node(
C:\Users\roysa\.conda\envs\udacity-genai\Lib\site-packages\diffusers\utils\outputs.py:63: UserWarning: torch.utils._pytree._register_pytree_node is deprecated. Please use torch.utils._pytree.register_pytree_node instead.
  torch.utils._pytree._register_pytree_node(

df

device = "cuda"
pipe = AutoPipelineForText2Image.from_pretrained(
    "stabilityai/sdxl-turbo", 
    torch_dtype=torch.float16, 
    variant="fp16"
).to(device)

Loading pipeline components...:   0%|          | 0/7 [00:00<?, ?it/s]

prompt = "A photo for real estate listing with the following property and neighborhood description:\n{}\n\n{}"
rand_gen = torch.manual_seed(9723947)

images = []
for i, row in df.iterrows():
    custom_prompt = prompt.format(row['description'], row['neighborhood_description'])
    image = pipe(
        prompt=custom_prompt, 
        num_inference_steps=3, # For this model you can use 1, but for normal Stable Diffusion you should use 25 or 50
        guidance_scale=1.0, # For this model 1 is fine, for normal Stable Diffusion you should use 6 or 7, or up to 10 or so
        negative_prompt=[], 
        generator=rand_gen
    ).images[0]
    images.append(image)

Token indices sequence length is longer than the specified maximum sequence length for this model (177 > 77). Running this sequence through the model will result in indexing errors
The following part of your input was truncated because CLIP can only handle sequences up to 77 tokens: ['- concept kitchen and dining area lead to a spacious backyard with a vegetable garden, perfect for the eco - conscious family. embrace sustainable living without compromising on style in this green oaks gem. green oaks is a close - knit, environmentally - conscious community with access to organic grocery stores, community gardens, and bike paths. take a stroll through the nearby green oaks park or grab a cup of coffee at the cozy green bean cafe. with easy access to public transportation and bike lanes, commuting is a breeze.']
Token indices sequence length is longer than the specified maximum sequence length for this model (177 > 77). Running this sequence through the model will result in indexing errors
The following part of your input was truncated because CLIP can only handle sequences up to 77 tokens: ['- concept kitchen and dining area lead to a spacious backyard with a vegetable garden, perfect for the eco - conscious family. embrace sustainable living without compromising on style in this green oaks gem. green oaks is a close - knit, environmentally - conscious community with access to organic grocery stores, community gardens, and bike paths. take a stroll through the nearby green oaks park or grab a cup of coffee at the cozy green bean cafe. with easy access to public transportation and bike lanes, commuting is a breeze.']

  0%|          | 0/3 [00:00<?, ?it/s]

C:\Users\roysa\.conda\envs\udacity-genai\Lib\site-packages\diffusers\models\attention_processor.py:1244: UserWarning: 1Torch was not compiled with flash attention. (Triggered internally at ..\aten\src\ATen\native\transformers\cuda\sdp_utils.cpp:263.)
  hidden_states = F.scaled_dot_product_attention(
The following part of your input was truncated because CLIP can only handle sequences up to 77 tokens: ['entertaining. enjoy the convenience of living within walking distance to trendy restaurants, art galleries, and boutiques. the downtown loft district is a bustling neighborhood known for its historic architecture, art scene, and lively nightlife. explore the nearby art galleries, attend live music events, or dine at the trendy restaurants lining the streets. with easy access to public transportation and major highways, commuting to work or exploring the city is a breeze.']
The following part of your input was truncated because CLIP can only handle sequences up to 77 tokens: ['entertaining. enjoy the convenience of living within walking distance to trendy restaurants, art galleries, and boutiques. the downtown loft district is a bustling neighborhood known for its historic architecture, art scene, and lively nightlife. explore the nearby art galleries, attend live music events, or dine at the trendy restaurants lining the streets. with easy access to public transportation and major highways, commuting to work or exploring the city is a breeze.']

  0%|          | 0/3 [00:00<?, ?it/s]

The following part of your input was truncated because CLIP can only handle sequences up to 77 tokens: ['retreat with a luxurious en - suite bathroom and walk - in closet. suburban meadows is a quiet and safe neighborhood with top - rated schools, parks, and community events. enjoy the convenience of shopping centers, restaurants, and recreational facilities nearby. with easy access to highways and public transportation, commuting to work or exploring the city is a breeze.']
The following part of your input was truncated because CLIP can only handle sequences up to 77 tokens: ['retreat with a luxurious en - suite bathroom and walk - in closet. suburban meadows is a quiet and safe neighborhood with top - rated schools, parks, and community events. enjoy the convenience of shopping centers, restaurants, and recreational facilities nearby. with easy access to highways and public transportation, commuting to work or exploring the city is a breeze.']

  0%|          | 0/3 [00:00<?, ?it/s]

The following part of your input was truncated because CLIP can only handle sequences up to 77 tokens: ['convenience of living within walking distance to boutique shops, cafes, and historic landmarks. old town is a picturesque neighborhood known for its historic architecture, cobblestone streets, and quaint shops. explore the nearby museums, art galleries, and cultural events, or take a leisurely stroll along the riverfront. with easy access to public transportation and bike paths, commuting to work or exploring the city is a breeze.']
The following part of your input was truncated because CLIP can only handle sequences up to 77 tokens: ['convenience of living within walking distance to boutique shops, cafes, and historic landmarks. old town is a picturesque neighborhood known for its historic architecture, cobblestone streets, and quaint shops. explore the nearby museums, art galleries, and cultural events, or take a leisurely stroll along the riverfront. with easy access to public transportation and bike paths, commuting to work or exploring the city is a breeze.']

  0%|          | 0/3 [00:00<?, ?it/s]

The following part of your input was truncated because CLIP can only handle sequences up to 77 tokens: ['for entertaining guests while enjoying the sunset over the water. waterfront harbor is a prestigious waterfront community with luxury homes, marinas, and yacht clubs. enjoy the convenience of living near upscale restaurants, boutiques, and waterfront parks. with easy access to major highways and private boat docks, commuting to work or exploring the city by water is a breeze.']
The following part of your input was truncated because CLIP can only handle sequences up to 77 tokens: ['for entertaining guests while enjoying the sunset over the water. waterfront harbor is a prestigious waterfront community with luxury homes, marinas, and yacht clubs. enjoy the convenience of living near upscale restaurants, boutiques, and waterfront parks. with easy access to major highways and private boat docks, commuting to work or exploring the city by water is a breeze.']

  0%|          | 0/3 [00:00<?, ?it/s]

idx = 11
print('Description:', df.iloc[idx]['description'])
print('Neighborhood Description:', df.iloc[idx]['neighborhood_description'])
images[idx]

Description: Welcome to this spacious 4-bedroom, 3-bathroom home in the family-friendly Suburb neighborhood. The open floor plan features a cozy fireplace in the living room, a gourmet kitchen with granite countertops, and a backyard oasis with a pool and patio for outdoor entertaining. The master suite offers a private retreat with a luxurious en-suite bathroom and walk-in closet.
Neighborhood Description: Suburb is a quiet and safe neighborhood with top-rated schools, parks, and community events. Enjoy the convenience of shopping centers, restaurants, and recreational facilities nearby. With easy access to highways and public transportation, commuting to work or exploring the city is a breeze.

idx = 13
print('Description:', df.iloc[idx]['description'])
print('Neighborhood Description:', df.iloc[idx]['neighborhood_description'])
images[idx]

Description: Step back in time with this charming 3-bedroom, 2-bathroom home in the historic District neighborhood. The home features original hardwood floors, crown molding, and a cozy fireplace in the living room. The updated kitchen boasts stainless steel appliances and a breakfast nook overlooking the landscaped backyard. Enjoy the convenience of living within walking distance to boutique shops, cafes, and historic landmarks.
Neighborhood Description: District is a picturesque neighborhood known for its historic architecture, cobblestone streets, and quaint shops. Explore the nearby museums, art galleries, and cultural events, or take a leisurely stroll along the riverfront. With easy access to public transportation and bike paths, commuting to work or exploring the city is a breeze.

import os
images_dir = 'images'

# Create image directory
if not os.path.exists(images_dir):
    os.makedirs(images_dir)
# Save each image
for i, image in enumerate(images):
    filename = os.path.join(images_dir, str(i) + ".png")
    image.save(filename)

import pandas as pd
import os
import PIL

images_dir = 'images'
images = []
text_template = \
"""
Description: {}
Neighborhood Description: {}
Price: {}
Bedrooms: {}
Bathrooms: {}
House Size (sqft): {}
"""

df = pd.read_csv('listings.csv')
image_paths = []
texts = []
ids = [{'id':i} for i in range(len(df.index))]

for i, row in df.iterrows():
    image_file = os.path.join(images_dir, str(i) + ".png")
    images.append(PIL.Image.open(image_file))
    image_paths.append(image_file)
    texts.append(text_template.format(row['description'],
                               row['neighborhood_description'],
                               row['price'],
                               row['bedrooms'],
                               row['bathrooms'],
                               row['house_size']))

from langchain_experimental.open_clip import OpenCLIPEmbeddings
from langchain.vectorstores import Chroma

# Create vector store
db = Chroma(
    collection_name="listings", embedding_function=OpenCLIPEmbeddings()
)

# Add text information
db.add_texts(texts=texts, metadatas = ids)

# Add image information
db.add_images(uris=image_paths, metadatas = ids)

['8b0e196d-d596-11ee-a06e-bc091beacf78',
 '8b0e196e-d596-11ee-809b-bc091beacf78',
 '8b0e196f-d596-11ee-a782-bc091beacf78',
 '8b0e1970-d596-11ee-a110-bc091beacf78',
 '8b0e1971-d596-11ee-af43-bc091beacf78',
 '8b0e1972-d596-11ee-ad32-bc091beacf78',
 '8b0e1973-d596-11ee-a4ab-bc091beacf78',
 '8b0e1974-d596-11ee-b50a-bc091beacf78',
 '8b0e1975-d596-11ee-a1fc-bc091beacf78',
 '8b0e1976-d596-11ee-ac58-bc091beacf78',
 '8b0e1977-d596-11ee-b3d1-bc091beacf78',
 '8b0e1978-d596-11ee-b014-bc091beacf78',
 '8b0e1979-d596-11ee-a87a-bc091beacf78',
 '8b0e197a-d596-11ee-9c2d-bc091beacf78',
 '8b0e197b-d596-11ee-87dd-bc091beacf78']

res = db.similarity_search("", k=100)

WARNING:chromadb.segment.impl.vector.local_hnsw:Number of requested results 100 is greater than number of elements in index 30, updating n_results = 30

QUESTIONS = [   
                "How big do you want your house to be?",
                "What are 3 most important things for you in choosing this property?", 
                "Which amenities would you like?", 
                "Which transportation options are important to you?",
                "How urban do you want your neighborhood to be?",   
            ]

def get_top_listings(preferences: list[str], top_k: int=5) -> list[int]:
    '''
    Returns the ids of the top k listing 
    '''
    top_listings = []
    combined_pref = '\n'.join(preferences)
    
    result = db.similarity_search(combined_pref, k=top_k*2)
    for res in result:
        # skip if it's image r text counterpart is already considered
        # in the ranking
        if res.metadata['id'] in top_listings:
            continue
        top_listings.append(res.metadata['id'])
        
        # already gotten top listings
        if len(top_listings) == top_k:
            break
    return top_listings

ANSWERS = [
                "A comfortable three-bedroom house with a spacious kitchen and a cozy living room.",
                "A quiet neighborhood, good local schools, and convenient shopping options.",
                "A backyard for gardening, a two-car garage, and a modern, energy-efficient heating system.",
                "Easy access to a reliable bus line, proximity to a major highway, and bike-friendly roads.",
                "A balance between suburban tranquility and access to urban amenities like restaurants and theaters."
          ]
res = get_top_listings(ANSWERS)

# check top listing
print(texts[res[0]])
images[res[0]]

Description: Step back in time with this charming 3-bedroom, 2-bathroom home in the historic District neighborhood. The home features original hardwood floors, crown molding, and a cozy fireplace in the living room. The updated kitchen boasts stainless steel appliances and a breakfast nook overlooking the landscaped backyard. Enjoy the convenience of living within walking distance to boutique shops, cafes, and historic landmarks.
Neighborhood Description: District is a picturesque neighborhood known for its historic architecture, cobblestone streets, and quaint shops. Explore the nearby museums, art galleries, and cultural events, or take a leisurely stroll along the riverfront. With easy access to public transportation and bike paths, commuting to work or exploring the city is a breeze.
Price: 900000
Bedrooms: 3
Bathrooms: 2
House Size (sqft): 2200

ANSWERS = [
                "A cozy living room.",
                "Huge windows, Furnished with a nice grey sofa for stylish comfort with some lamp shade",
                "",
                "",
                ""
          ]
res = get_top_listings(ANSWERS)

# check top listing
print(texts[res[0]])
images[res[0]]

Description: Live in style in this modern 3-bedroom, 2-bathroom loft in the trendy Urban Loft District. The open floor plan features high ceilings, concrete floors, and industrial-chic finishes. The gourmet kitchen is equipped with stainless steel appliances and a breakfast bar. Enjoy the convenience of living within walking distance to hip cafes, art galleries, and boutique shops.
Neighborhood Description: The Urban Loft District is a hip and trendy neighborhood known for its converted industrial spaces, art scene, and creative community. Explore the nearby galleries, attend art events, or dine at the trendy restaurants lining the streets. With easy access to public transportation and bike lanes, commuting to work or exploring the city is a breeze.
Price: 1000000
Bedrooms: 3
Bathrooms: 2
House Size (sqft): 1900

from langchain.llms import OpenAI
from langchain.prompts import PromptTemplate

model_name = 'gpt-3.5-turbo'
llm = OpenAI(model_name=model_name, temperature=0)

INSTRUCTION = \
"""
Create a short statement for each listing to entice the buyer given their preferences. Try to paraphrase and highlight what the buyer wants.
"""

C:\Users\roysa\.conda\envs\udacity-genai\Lib\site-packages\langchain_community\llms\openai.py:249: UserWarning: You are trying to use a chat model. This way of initializing it is no longer supported. Instead, please use: `from langchain_community.chat_models import ChatOpenAI`
  warnings.warn(
C:\Users\roysa\.conda\envs\udacity-genai\Lib\site-packages\langchain_community\llms\openai.py:1070: UserWarning: You are trying to use a chat model. This way of initializing it is no longer supported. Instead, please use: `from langchain_community.chat_models import ChatOpenAI`
  warnings.warn(

from langchain.output_parsers import PydanticOutputParser
from langchain_core.pydantic_v1 import BaseModel, Field, NonNegativeInt
from typing import List

class PersonalizedSummary(BaseModel):
    id: int = Field(description="ID of the listing")
    summary: str = Field(description="Generated personalized summaries")

class SummaryCollection(BaseModel):
    summaries: List[PersonalizedSummary] = Field(description="List of generated personalized summaries")
        
parser = PydanticOutputParser(pydantic_object=SummaryCollection)
prompt = PromptTemplate(
    template="{instruction}\nBuyer Preferences:\n{buyer_preferences}\nListings:\n{listings}\n\n{format_instructions}",
    input_variables=["instruction", "buyer_preferences", "listings"],
    partial_variables={"format_instructions": parser.get_format_instructions},
)

cached_summary = {}

def personalize_top_listings(preferences: list[str], top_k: int=5) -> list:
    '''
    Returns a dictionary of personalized description of the listings based
    in user preference
    '''
    # Get the top related listings first
    res = get_top_listings(preferences, top_k)
    top_texts = []
    for idx in res:
        top_texts.append(''.join(['ID:' + str(idx), texts[idx]]))
        
    query = prompt.format(instruction = INSTRUCTION, 
                          buyer_preferences = '\n'.join(preferences),
                          listings = '\n'.join(top_texts))
    output = []
    for summary in parser.parse(llm(query)).summaries:
        output.append((summary.id,summary.summary))
        cached_summary[summary.id] = summary.summary
    return output

ANSWERS = [
                "A cozy living room.",
                "Huge windows, Furnished with a nice grey sofa for stylish comfort with some lamp shade",
                "",
                "",
                ""
          ]
ans = personalize_top_listings(ANSWERS)
ans

[(10,
  'Live in style in this modern 3-bedroom, 2-bathroom loft in the trendy Urban Loft District. The open floor plan features high ceilings, concrete floors, and industrial-chic finishes. The gourmet kitchen is equipped with stainless steel appliances and a breakfast bar. Enjoy the convenience of living within walking distance to hip cafes, art galleries, and boutique shops.'),
 (5,
  'Live in the heart of the vibrant Arts District in this modern 2-bedroom, 2-bathroom condo. The open floor plan features high ceilings, hardwood floors, and floor-to-ceiling windows offering city views. The gourmet kitchen is equipped with stainless steel appliances and a breakfast bar. Enjoy the convenience of living within walking distance to art galleries, theaters, and trendy restaurants.'),
 (1,
  'Luxury living in the heart of the vibrant Downtown Loft District. This modern 2-bedroom, 2-bathroom loft features high ceilings, exposed brick walls, and floor-to-ceiling windows offering stunning city views. The sleek kitchen is equipped with top-of-the-line appliances and a spacious island for entertaining. Enjoy the convenience of living within walking distance to trendy restaurants, art galleries, and boutiques.'),
 (14,
  'Escape to the tranquil Lakeview Retreat in this spacious 5-bedroom, 4-bathroom home. The home features a grand foyer, formal living and dining rooms, and a gourmet kitchen with lake views. The master suite offers a private balcony overlooking the water, a spa-like bathroom, and a walk-in closet. The backyard oasis includes a pool, spa, and lush landscaping for outdoor entertaining.'),
 (11,
  'Welcome to this spacious 4-bedroom, 3-bathroom home in the family-friendly Suburb neighborhood. The open floor plan features a cozy fireplace in the living room, a gourmet kitchen with granite countertops, and a backyard oasis with a pool and patio for outdoor entertaining. The master suite offers a private retreat with a luxurious en-suite bathroom and walk-in closet.')]

ANSWERS = [
                "A cozy living room.",
                "Mobility and traveling is important for me",
                "",
                "",
                ""
          ]
ans = personalize_top_listings(ANSWERS)
ans

[(8,
  'Experience historic charm in this 3-bedroom, 2-bathroom home in the heart of Historic Downtown. Original hardwood floors, exposed brick walls, and a cozy fireplace in the living room. Updated kitchen with stainless steel appliances. Walking distance to historic landmarks, museums, and trendy restaurants.'),
 (10,
  'Live in style in this modern 3-bedroom, 2-bathroom loft in the trendy Urban Loft District. High ceilings, concrete floors, and industrial-chic finishes. Gourmet kitchen with stainless steel appliances. Walking distance to hip cafes, art galleries, and boutique shops.'),
 (3,
  'Step back in time with this charming 3-bedroom, 2-bathroom home in the historic Old Town neighborhood. Original hardwood floors, crown molding, and a cozy fireplace in the living room. Updated kitchen with stainless steel appliances. Walking distance to boutique shops, cafes, and historic landmarks.'),
 (13,
  'Step back in time with this charming 3-bedroom, 2-bathroom home in the historic District neighborhood. Original hardwood floors, crown molding, and a cozy fireplace in the living room. Updated kitchen with stainless steel appliances. Walking distance to boutique shops, cafes, and historic landmarks.'),
 (11,
  'Welcome to this spacious 4-bedroom, 3-bathroom home in the family-friendly Suburb neighborhood. Cozy fireplace in the living room, gourmet kitchen with granite countertops. Backyard oasis with a pool and patio for outdoor entertaining. Master suite with luxurious en-suite bathroom and walk-in closet.')]

import gradio as gr

# Auxiliary event handling methods
def set_sr_tab(request: gr.Request):
    return gr.Tabs(selected=1)

def get_select_index(evt: gr.SelectData):
    listing_id = int(evt._data['value']['image']['orig_name'].split('.')[0])
    return cached_summary[listing_id]

# Fetch personalized listings from user preference
def get_search_result(state, answer0, answer1, answer2, answer3, answer4):
    state = 1
    #ans = personalize_top_listings([answer0, answer1, answer2, answer3, answer4])
    search_results = []
    for id, summary in personalize_top_listings([answer0, answer1, answer2, answer3, answer4]):
        search_results.append((image_paths[id], df.iloc[id]['neighborhood']))
    return [gr.update(selected=state), state, search_results]


# Draw layout of the Graphical User Interface
with gr.Blocks() as demo:
    label = gr.Label("My HomeMatch Application")

    with gr.Tabs() as tabs:
        state = gr.State(0)
        with gr.TabItem("Buyer's Preference", id=0):
            # Create textbox for each question for free-form response
            answers = [] 
            for i in range(len(QUESTIONS)):
                answers.append(gr.Textbox(label=QUESTIONS[i]))
            btn = gr.Button("Submit", scale=0)
    
        # Search results panel
        with gr.TabItem("Search Results", id=1):
            with gr.Row():
                with gr.Column():
                    gallery = gr.Gallery(
                        label="Recommended For You", show_label=True, elem_id="gallery", 
                        columns=[3], rows=[2], object_fit="contain", height="auto", 
                        allow_preview=True)
        
                with gr.Column():
                    personalized_desc = gr.Label(label="Personalized Description")
                    #personalized_desc = gr.Label(label="Original Description")
             
    
    # Set event handling
    btn.click(fn=get_search_result, 
              inputs=[state] + [answers[i] for i in range(len(QUESTIONS))], 
              outputs=[tabs, state, gallery])

    gallery.select(get_select_index, 
                   None, 
                   outputs=personalized_desc)

if __name__ == "__main__":
    demo.launch()

Running on local URL:  http://127.0.0.1:7874

To create a public link, set `share=True` in `launch()`.

	neighborhood	price	bedrooms	bathrooms	house_size	description	neighborhood_description
0	Green Oaks	800000	3	2	2000	Welcome to this eco-friendly oasis nestled in ...	Green Oaks is a close-knit, environmentally-co...
1	Downtown Loft District	1200000	2	2	1800	Luxury living in the heart of the vibrant Down...	The Downtown Loft District is a bustling neigh...
2	Suburban Meadows	600000	4	3	2500	Welcome to this spacious 4-bedroom, 3-bathroom...	Suburban Meadows is a quiet and safe neighborh...
3	Historic Old Town	950000	3	2	2200	Step back in time with this charming 3-bedroom...	Old Town is a picturesque neighborhood known f...
4	Waterfront Harbor	1500000	5	4	3500	Experience waterfront living in this stunning ...	Waterfront Harbor is a prestigious waterfront ...

	id	neighborhood	price	bedrooms	bathrooms	house_size	description	neighborhood_description
0	0	Green Oaks	800000	3	2	2000	Welcome to this eco-friendly oasis nestled in ...	Green Oaks is a close-knit, environmentally-co...
1	1	Downtown Loft District	1200000	2	2	1800	Luxury living in the heart of the vibrant Down...	The Downtown Loft District is a bustling neigh...
2	2	Suburban Meadows	600000	4	3	2500	Welcome to this spacious 4-bedroom, 3-bathroom...	Suburban Meadows is a quiet and safe neighborh...
3	3	Historic Old Town	950000	3	2	2200	Step back in time with this charming 3-bedroom...	Old Town is a picturesque neighborhood known f...
4	4	Waterfront Harbor	1500000	5	4	3500	Experience waterfront living in this stunning ...	Waterfront Harbor is a prestigious waterfront ...
5	5	Arts District	700000	2	2	1600	Live in the heart of the vibrant Arts District...	The Arts District is a cultural hub known for ...
6	6	Mountain View Estates	1200000	4	3	2800	Escape to the serene Mountain View Estates in ...	Mountain View Estates is a prestigious neighbo...
7	7	Beachfront Paradise	2000000	6	5	4000	Live the beachfront lifestyle in this luxuriou...	Beachfront Paradise is a prestigious beachfron...
8	8	Historic Downtown	850000	3	2	2100	Experience historic charm in this 3-bedroom, 2...	Historic Downtown is a vibrant neighborhood kn...
9	9	Lakefront Retreat	1800000	5	4	3500	Escape to the tranquil Lakefront Retreat in th...	Lakefront Retreat is a prestigious lakefront c...
10	10	Urban Loft District	1000000	3	2	1900	Live in style in this modern 3-bedroom, 2-bath...	The Urban Loft District is a hip and trendy ne...
11	11	Family-Friendly Suburb	700000	4	3	2400	Welcome to this spacious 4-bedroom, 3-bathroom...	Suburb is a quiet and safe neighborhood with t...
12	12	Riverside Haven	1300000	4	3	2700	Escape to the serene Riverside Haven in this s...	Riverside Haven is a prestigious neighborhood ...
13	13	Historic District	900000	3	2	2200	Step back in time with this charming 3-bedroom...	District is a picturesque neighborhood known f...
14	14	Lakeview Retreat	1600000	5	4	3300	Escape to the tranquil Lakeview Retreat in thi...	Lakeview Retreat is a prestigious lakefront co...

Project: HomeMatch Generative AI Application¶

Prerequisites¶

Listing Generation¶

Image Generation¶

Multimodal Vector Store and Embeddings¶

Buyer's Preference and Semantic Search¶

Personalization¶

User Interface¶

Running The Prototype Application¶